KrauthammerLab
University of Zurich & University Hospital Zurich

Predicting base editing efficiencies in both lab cells and living models

Generative DL models patient trajectories

Studying how conversational coordination shapes meaning in human and machine communication

Modeling speech and behavior to support mental health research and assessment

Audio-first approaches to understanding language directly from speech

Generative AI for protein fitness optimization

Helping reduce idle time in the USZ Radiology department
This study investigates the use of cfDNA sequencing to monitor tumor dynamics in patients undergoing high-dose radiotherapy, revealing correlations between genetic alterations and clinical outcomes.
We developed machine learning models, PRIDICT2.0 and ePRIDICT, to predict prime editing efficiency, offering a robust tool for optimizing genome editing strategies across diverse chromatin contexts.
The first objective of this study was to implement and assess the performance and reliability of a vision transformer (ViT)-based deep-learning model, an ‘off-the-shelf’ artificial intelligence solution, for identifying distinct signs of microangiopathy in nailfold capilloroscopy (NFC) images of patients with SSc. The second objective was to compare the ViT’s analysis performance with that of practising rheumatologists.